Regulation of hepatic redox state plays a central role in normal physiology and high-impact diseases including non-alcoholic fatty liver disease and hepatotoxicity induced by many drugs. Liver biopsy is recommended for patients with those problems to monitor hepatic oxidative stress and other progress of the diseases, but it is invasive. There is a demand for a simple, less invasive method to monitor redox disruption in many hepatic diseases. It would benefit to develop a method to assess flux in key reactions involved in the response to oxidative stress and protection of the redox state of the liver. The pentose phosphate pathway (PPP) produces NADPH reducing equivalent that is essential for antioxidant defense and it has intricate interactions with hepatic glucose production. Recently we found that the liver given [U- 13C3]glycerol produced unique 13C asymmetry between carbons 1-3 and 4-6 of blood glucose, which was sensitive to the PPP and transaldolase activity. Transaldolase, a key enzyme in the non-oxidative phase of the PPP, is considered to have roles under oxidative stress. Here we hypothesize that the amounts of 13C glucose isotopomers produced and released from the liver given [U-13C3]glycerol are sensitive to hepatic oxidative stress. To test this, we will determine correlations in the following events: acetaminophen stressed or fatty liver -> hepatic oxidative stress -> altered hepatic PPP and transaldolase activity -> altered 13C asymmetry of glucose released from the liver. Both rodent models and human subjects with fatty liver will be studied to determine 13C asymmetry of glucose caused by chronic oxidative stress. We will determine if the asymmetry from patients with nonalcoholic steatohepatitis can be distinguished from simple steatosis. In addition, we will study rodent models of acetaminophen-induced hepatotoxicity and healthy subjects given therapeutic doses of acetaminophen if 13C asymmetry of glucose is sensitive to the drug. The results will be compared with gene expression data and standard methods for oxidative stress measurement. The overall procedures are simple: [U-13C3]glycerol ingestion, blood draw and blood glucose analysis by 13C NMR. Liver fat content in human subjects will be measured using 1H MRS. Our goal is to develop a simple method to identify and monitor hepatic oxidative stress by observing 13C asymmetry in blood glucose, which is readily determined from a single blood sample.